patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Jerin Jacob <jerinjacobk@gmail.com>
To: Gaoxiang Liu <gaoxiangliu0@163.com>
Cc: dpdk-dev <dev@dpdk.org>,
	Anatoly Burakov <anatoly.burakov@intel.com>,
	liugaoxiang@huawei.com, dpdk stable <stable@dpdk.org>
Subject: Re: [PATCH] mem: exclude used memory from core dump by config
Date: Tue, 14 Dec 2021 17:38:51 +0530	[thread overview]
Message-ID: <CALBAE1M=Kqrc7R8_N30gJqLLefyM-tNXhoBPNVDHa-_j_Bh2eA@mail.gmail.com> (raw)
In-Reply-To: <20211214114151.1298-1-gaoxiangliu0@163.com>

On Tue, Dec 14, 2021 at 5:12 PM Gaoxiang Liu <gaoxiangliu0@163.com> wrote:
>
> Some DPDK application is allolcateed storage partition of 8G(or smaller)

allocated

> If coredump happens, the application doesn't work because of
> insufficient storage space.
> The patch provides a config that means whether the huge used memory
> needs to be not dumped in core dump.
> The DPDK application can choose to open it according to the actual
> situation.
>
> Fixes: d72e4042c5eb ("mem: exclude unused memory from core dump")
> Cc: stable@dpdk.org

Probably subject can be changed to: eal: allow to exclude memseg from core dump

>
> Signed-off-by: Gaoxiang Liu <liugaoxiang@huawei.com>
> ---
>  lib/eal/common/eal_internal_cfg.h |  1 +
>  lib/eal/common/eal_options.h      |  3 ++-
>  lib/eal/linux/eal.c               |  3 +++
>  lib/eal/linux/eal_memalloc.c      | 10 ++++++++++

freebsd support can be updated,
Please update doc/guides/linux_gsg/linux_eal_parameters.rst and
doc/guides/freebsd_gsg/freebsd_eal_parameters.rst.


>  4 files changed, 16 insertions(+), 1 deletion(-)
>
> diff --git a/lib/eal/common/eal_internal_cfg.h b/lib/eal/common/eal_internal_cfg.h
> index d6c0470eb8..f9cd300f86 100644
> --- a/lib/eal/common/eal_internal_cfg.h
> +++ b/lib/eal/common/eal_internal_cfg.h
> @@ -87,6 +87,7 @@ struct internal_config {
>                         /**< user defined mbuf pool ops name */
>         unsigned num_hugepage_sizes;      /**< how many sizes on this system */
>         struct hugepage_info hugepage_info[MAX_HUGEPAGE_SIZES];
> +       volatile unsigned huge_dont_dump_flag;

Volatile may not be needed. it can be changed to bool as well.


>         enum rte_iova_mode iova_mode ;    /**< Set IOVA mode on this system  */
>         rte_cpuset_t ctrl_cpuset;         /**< cpuset for ctrl threads */
>         volatile unsigned int init_complete;
> diff --git a/lib/eal/common/eal_options.h b/lib/eal/common/eal_options.h
> index 8e4f7202a2..6b9a24f0f9 100644
> --- a/lib/eal/common/eal_options.h
> +++ b/lib/eal/common/eal_options.h
> @@ -87,7 +87,8 @@ enum {
>         OPT_NO_TELEMETRY_NUM,
>  #define OPT_FORCE_MAX_SIMD_BITWIDTH  "force-max-simd-bitwidth"
>         OPT_FORCE_MAX_SIMD_BITWIDTH_NUM,
> -
> +#define OPT_HUGE_DONT_DUMP     "huge-dont-dump"
> +       OPT_HUGE_DONT_DUMP_NUM,
>         OPT_LONG_MAX_NUM
>  };
>
> diff --git a/lib/eal/linux/eal.c b/lib/eal/linux/eal.c
> index 60b4924838..eeab5e4f3d 100644
> --- a/lib/eal/linux/eal.c
> +++ b/lib/eal/linux/eal.c
> @@ -816,6 +816,9 @@ eal_parse_args(int argc, char **argv)
>                 case OPT_MATCH_ALLOCATIONS_NUM:
>                         internal_conf->match_allocations = 1;
>                         break;
> +               case OPT_HUGE_DONT_DUMP_NUM:
> +                       internal_config->huge_dont_dump_flag = 1;
> +                       break;
>
>                 default:
>                         if (opt < OPT_LONG_MIN_NUM && isprint(opt)) {
> diff --git a/lib/eal/linux/eal_memalloc.c b/lib/eal/linux/eal_memalloc.c
> index 337f2bc739..13ca1cf58c 100644
> --- a/lib/eal/linux/eal_memalloc.c
> +++ b/lib/eal/linux/eal_memalloc.c
> @@ -663,6 +663,16 @@ alloc_seg(struct rte_memseg *ms, void *addr, int socket_id,
>         ms->iova = iova;
>         ms->socket_id = socket_id;
>
> +       if (internal_config->huge_dont_dump_flag) {
> +               if (madvise(addr, alloc_sz, EAL_DONTDUMP)) {
> +                       RTE_LOG(INFO, EAL, "madvise(%p, %#zx, %d) failed: %s\n",
> +                                       virt, size, flags, strerror(rte_errno));
> +               } else {
> +                       RTE_LOG(INFO, EAL, "%s(): madvise(MADV_DONTDUMP) memory success.(len=%zu)\n",
> +                                       __func__, alloc_sz);
> +               }

I think, you can use eal_mem_set_dump()

> +       }
> +
>         return 0;
>
>  mapped:
> --
> 2.32.0
>
>

  parent reply	other threads:[~2021-12-14 12:09 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-14 11:41 Gaoxiang Liu
2021-12-14 12:08 ` [PATCH v2] " Gaoxiang Liu
2021-12-14 15:18   ` [PATCH v3] eal: allow to exclude memseg from core dump Gaoxiang Liu
2021-12-15 12:06     ` Jerin Jacob
2021-12-15 14:34       ` Dmitry Kozlyuk
2021-12-15 15:46     ` Dmitry Kozlyuk
2021-12-14 12:08 ` Jerin Jacob [this message]
2021-12-14 12:21   ` [PATCH] mem: exclude used memory from core dump by config Gaoxiang Liu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CALBAE1M=Kqrc7R8_N30gJqLLefyM-tNXhoBPNVDHa-_j_Bh2eA@mail.gmail.com' \
    --to=jerinjacobk@gmail.com \
    --cc=anatoly.burakov@intel.com \
    --cc=dev@dpdk.org \
    --cc=gaoxiangliu0@163.com \
    --cc=liugaoxiang@huawei.com \
    --cc=stable@dpdk.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).